lammps-16Feb16 compile error whit user cuda

Deal all,

Trying to compile lammps-16Feb16 whit user-cuda an error comes up (But whit out cuda it works),

RECISION=2 -DCUDA_ARCH=20 -c …/verlet_cuda.cpp
…/verlet_cuda.cpp: In member function ‘virtual void LAMMPS_NS::VerletCuda::setup()’:
…/verlet_cuda.cpp:126:17: error: ‘class LAMMPS_NS::Update’ has no member named ‘max_wall’
if (update->max_wall > 0) {
^
…/verlet_cuda.cpp:128:35: error: ‘class LAMMPS_NS::Update’ has no member named ‘max_wall’
double totalclock = update->max_wall;
^
…/verlet_cuda.cpp: In member function ‘virtual void LAMMPS_NS::VerletCuda::run(int)’:
…/verlet_cuda.cpp:658:17: error: ‘class LAMMPS_NS::Update’ has no member named ‘time_expired’
if (update->time_expired()) {
^
make[1]: *** [verlet_cuda.o] Error 1
make[1]: Leaving directory `/home/fluid_reza/lammps-16Feb16/src/Obj_ompi_g++’
make: *** [ompi_g++] Error 2

Could you please help me?!

Thanks in Advance,

Hi,

Yes, i had similar errors in stable version 7Dec15; see below.
This seems to have been first reported in November:
https://sourceforge.net/p/lammps/mailman/message/34599351/

I recommend that this be added to
http://lammps.sandia.gov/unbug.html

It would also be useful to have a KNOWN_BUGS file in the distribution;
this might just contain the unbug link (if that is really active) or
links to the mailing list since that might be easier to manage than
actual textual descriptions in the file.

A known bugs file would have saved me some time.
One could spend days surfing the lammps web pages.
There are some LAMMPS installers that are not LAMMPS users.
I have some more suggestions that I've been planning to send ...

=== details
mpicc -O3 -xHost -fno-alias -ansi-alias -restrict -DLAMMPS_GZIP -DLAMMPS_JPEG -I../../lib/qmmm -DLMP_USER_OMP -I../../lib/cuda -DLMP_USER_CUDA -I../../lib/colvars -I../../lib/awpmd/ivutils/include -I../../lib/awpmd/systems/interact -I../../lib/atc -I../../lib/reax -DLMP_PYTHON -I../../lib/poems -DLMP_MPIIO -I../../lib/meam -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -DFFT_MKL -I/usr/local/cuda/7.5.18/include -DUNIX -DFFT_CUFFT -DCUDA_PRECISION=2 -DCUDA_ARCH=20 -I/usr/include/python2.6 -I/usr/include/python2.6 -c ../verlet_cuda.cpp
../verlet_cuda.cpp(126): error: class "LAMMPS_NS::Update" has no member "max_wall"
.
.
.
Intel(R) C Intel(R) 64 Compiler XE for applications running on Intel(R) 64, Version 15.0.3.187 Build 20150407
https://www.osc.edu/supercomputing/computing/oakley
https://www.osc.edu/supercomputing/computing/ruby
=== details

scott

Hi scott,

Thank you for your reply, what did you for your problem?

Reza,

Hi scott,

Thank you for your reply, what did you for your problem?

the error is due to some planned feature for LAMMPS, that accidentally
slipped in some time in october 2015. this has been corrected in the
svn and git repos yesterday.

you can thus download the updated file from the svn or git repo or
from this URL: http://git.lammps.org/git/?p=lammps-ro.git;a=blob_plain;f=src/USER-CUDA/verlet_cuda.cpp;hb=HEAD

axel.

Stan fixed this yesterday and it will be

in the next LAMMPS patch.

Steve

Thank you all.

Reza